**3. Results**

Sixteen patients aged between 16 and 59 years were enrolled. Basal characteristics of the included cohort are reported in Table 1. Most patients were female (*n* = 10/16, 38%). Patients mainly had normal weight (BMI < 25 kg/m2) and all of them consumed a Mediterranean diet [36]. No dietary changes were made prior to study initiation. According to FBDSI score, severity of IBS was mild in 1 patient (6%), moderate in 8 patients (50%), and severe in 7 patients (44%). Anamnestic collection showed that 4 of the 16 enrolled patients (25%) had lactose intolerance confirmed by hydrogen breath test (HBT). All a ffected patients had reported little or no benefit after the elimination of lactose from diet in the previous years, and no changes in the diet were performed during the study period.


**Table 1.** Baseline characteristics of the included patients.

Abbreviations: body mass index (BMI), confidence interval (CI), Functional Bowel Disorder Severity Index (FBDSI), interleukin (IL), Irritable Bowel Syndrome Quality of Life (IBS-QoL), Irritable Bowel Syndrome Severity Scoring System (IBS-SSS), *n* (number), standard deviation (SD), tumour necrosis factor-alpha (TNFα).

As expected, we observed a positive correlation between FBDSI and IBS-SSS (*rs* = 0.658, 95% CI 0.241–0.870, *p* = 0.006) and an inverse correlation between IBS-SSS and overall IBS-QoL score (*rs* = −0.550, 95% CI −0.822–−0.074, *p* = 0.027). Age positively correlated with serum zonulin levels (*rs* = 0.558, 95% CI 0.086–0.825, *p* = 0.025) and negatively correlated with FBDSI (*rs* = −0.570, 95% CI −0.831–−0.104, *p* = 0.021). A significant positive correlation was also observed between abdominal pain intensity and TNF α levels (*rs* = 0.585, 95% CI 0.126–0.838, *p* = 0.017). Only a slight positive trend was observed regarding the correlation between serum zonulin and BMI (*rs* = 0.459, 95% CI −0.047–0.778, *p* = 0.074), between serum zonulin and IL-6 values (*rs* = 0.453, 95% CI −0.055–0.775, *p* = 0.078) and between IL-8 and TNF α values (*rs* = 0.439, 95% CI −0.072–0.768, *p* = 0.089). No other significant correlations were observed between serum zonulin or cytokines levels and other demographic and clinical characteristics of the study population (Figure 1).

**Figure 1.** Correlation between age and serum zonulin levels (**A**), age and FBDSI (**B**), abdominal pain intensity and TNFα values (**C**), serum zonulin levels and BMI (**D**), serum zonulin and IL-6 values (**E**) and IL-8 and TNFα values (**F**). Abbreviations: Body Mass Index (BMI), Functional Bowel Disorder Severity Index (FBDSI), interleukin (IL), tumour necrosis factor-alpha (TNFα).

At baseline, no significant difference was observed between patients undergoing 8 weeks and 12 weeks of treatment regarding severity of disease, serum zonulin levels and cytokines values (Table 2).

**Table 2.** Comparison of disease severity, IBS-QoL overall score, Bristol Stool Chart, serum zonulin levels and cytokines values, assessed at baseline (T0), between the two groups of patients treated with *B. Longum* ES1 for 8 or 12 weeks.


Abbreviations: interleukin (IL), confidence interval (CI), number (*n*), non-responder (NR), responder (R), standard deviation (SD), tumour necrosis factor-alpha (TNFα), week (w).

The changes of clinical parameters, serum zonulin and cytokines levels from baseline to end of therapy in the overall population of patients with IBS-D treated with *B. longum* ES1 are reported in Table 3.


**Table 3.** Changes of questionnaires scores, serum zonulin levels and cytokines values from T0 to T1 in the overall study population.

Abbreviations: confidence interval (CI), interleukin (IL), Irritable Bowel Syndrome Quality of Life (IBS-QoL), Irritable Bowel Syndrome Severity Scoring System (IBS-SSS), number (*n*), standard deviation (SD), tumour necrosis factor-alpha (TNFα), statistically significant (\*).

The clinical response to probiotic therapy, in terms of IBS-SSS score decrease of ≥50 points, was observed in 5/16 patients (31%), of whom 4/8 (50%) were treated for 12 weeks and 1/8 (13%) was treated for 8 weeks (*p* = 0.282); indeed, a trend towards IBS-SSS score reduction from T0 to T1 was found only in patients treated for 12 weeks (236 ± 67 vs. 189 ± 80, *p* = 0.072). An improvement ≥30% of the abdominal pain intensity was found in 5/16 (31%) patients, all of them treated for 12 weeks (5/8; 63%); consistently, patients treated for 12 weeks showed a significant reduction in the intensity of abdominal pain compared to those treated for 8 weeks (from 63 ± 28 to 34 ± 28, *p* = 0.020 and from 52 ± 30 to 62 ± 24, *p* = 0.132, respectively). Conversely, no significant improvement in bloating was reported (*p* = 0.340), independent of therapy duration. An overall IBS-QoL score increase of ≥14 points was obtained only in 1/16 (6%), who was treated for 12 weeks. Stool consistency improved regardless of the duration of therapy (*p* < 0.001). Nine out of 16 (56%) patients normalized stool consistency (type 3 or 4 according to Bristol Stool Scale): 5/8 (63%) were treated for 12 weeks, whereas 4/8 (50%) were treated for 8 weeks (*p* = 0.626). None of the patients reported adverse events during or after treatment.

Finally, we observed a significant decrease in IL-6, IL-8, IL-12 and TNFα values from baseline to end of therapy, irrespective of treatment duration. Conversely, no significant reduction was found in serum zonulin levels in the overall study population; however, in patients treated for 12 weeks, serum zonulin decreased significantly from 43.8 ± 6.8 ng/mL at T0 to 40.8 ± 5.0 ng/mL at T1 (*p* = 0.036) (Figure 2).

**Figure 2.** Variation of serum zonulin levels from baseline (T0) to end of therapy (T1) in patients with IBS-D treated with *B. longum* ES1 for 8 weeks or 12 weeks.
